The purpose of this proposal is to use mouse models to understand the role of TNF? and its receptors in the maintenance of B cell tolerance to lupus related autoantigens. TNF mediates both cell survival and cell death by binding to two receptors that are expressed by many cell types. TNFR1 mediates many inflammatory functions of TNF but also has a death domain, thereby allowing TNF to limit the inflammatory response by killing activated T cells. TNFR2 preferentially recognizes membrane TNF and may have a unique role in regulating costimulation of B cells and modulating the function of Tregs. TNF excess is associated with exacerbation of tissue injury in autoimmune diseases; this has been the basis for the successful use of TNF inhibition for inflammatory autoimmune diseases. Paradoxically, approximately 30% of patients treated with TNF inhibitors develop lupus related autoantibodies including anti-dsDNA and anti-cardiolipin antibodies and a significantly smaller proportion develops clinical lupus and other autoimmune diseases such as thromboses, psoriasis, MS and vasculitis. The mechanism by which TNF deficiency induces pathogenic autoantibodies is still not well understood, especially since TNF is required for the formation of germinal centers. We propose to study the mechanism for induction of lupus related autoantibodies in Sle1 mice bearing the site-directed autoreactive Ig heavy chain transgene 3H9 and deficient in TNF or its receptors. We have found that deficiency of either TNF or TNFR1 has profound effects on selection of both nave and activated B cells and induces a high titer anti-cardiolipin antibody response despite the absence of germinal centers. By contrast, deficiency of TNFR2 protects against the formation of autoantibodies despite normal germinal center formation. We hypothesize that TNF deficiency induces dysregulation of autoreactive B cells in extrafollicular regions resulting in the production of germline encoded low affinity antibodies. A second hit, either genetic or inflammatory is then required to induce somatically mutated autoantibodies that are pathogenic. We further hypothesize a role for membrane TNF as a costimulator of B cells expressing TNFR2. To address these hypotheses we propose two specific aims: the first will ask how deficiency of TNF and its receptors alters B cell tolerance by analyzing the positioning and selection of 3H9 B cells; the second will explore extrinsic factors that initiate autoantibody formation and pathogenic autoimmunity in TNF and TNFR1 deficient mice by inhibiting or activating innate and adaptive immunity. Given that millions of patients around the world are currently being treated with TNF inhibitors, autoimmune side effects are a significant medical problem and new approaches to prevention based on an understanding of the biologic mechanism that causes them are required. Our experiments will add a mechanistic understanding to how TNF inhibition induces autoimmunity and may yield approaches to the prevention of autoimmunity induced by TNF inhibitors, allowing safer use of these drugs.